**Figure 2.** **Autophagy genes can distinguish healthy and periodontitis samples.** (**A**) Univariate logistic regression investigated the relationship between dysregulated autophagy genes and periodontitis. (**B**) Least absolute shrinkage and selection operator (LASSO) coefficient profiles of 16 periodontitis-related autophagy genes. (**C**) Ten-fold cross-validation for tuning parameter selection in the LASSO regression. The partial likelihood deviance is plotted against log (λ), where λ is the tuning parameter. Partial likelihood deviance values are shown, with error bars representing SE. The dotted vertical lines are drawn at the optimal values by minimum criteria and 1-SE criteria. (**D**) Distinguishing signature with 10 autophagy genes was developed by multivariate logistic regression and the risk scores for periodontitis were calculated. (**E**) The risk distribution between healthy and periodontitis, where periodontitis has a much higher risk score than healthy samples. (**F**) Principal component analysis (PCA) of 10 periodontitis-related autophagy genes between healthy and periodontitis. The two first principal components (PC1, PC2) which could explain most of the variables are plotted. (**G**) The discrimination ability for healthy and periodontitis samples by autophagy genes was analyzed by the ROC curve and evaluated by AUC value.